This application uses human event-related potentials and reaction time to address several issues concerning linguistic processing. These include: 1) assessing the rate at which passive decay of automatic spreading activation occurs, using a paradigm devised by Neely to disentangle semantic priming due to expectations on the part of the subject and from that which is due to automatic spreading activation; 2) testing an hypothesis of Degenbach, Carr and Wihelmsen that under certain circumstances a center-surround attentional mechanism dampens activation that spreads from a prime to neighboring words; 3) testing the hypothesis that automatic spreading activation is terminated by the presentation of a word that is unrelated to a previous word; 4) obtaining electrophysiological evidence that would converge with RT data in supporting the hypothesis of Meyer, Schvanevelt and Ruddy that semantic priming may affect stimulus encoding; 5) locating the level of processing at which automatic spreading activation is initiated and the direction in which it spreads; 6) ascertaining whether semantic as well as lexical processing contributes to the elicitation of the N4OO component of human ERPs.